Techies in yet?
I do wonder how Techies will work in Dota 2 in very high MMR. When I played him in DotA, the average skill level was so much lower compared to what I play against now in AP. You could always get first blood on easy lane with well placed mines and people hardly ever bought a gem or sentries. Nowadays there are supports who constantly carry sentries if they are against something invisible.
i was catching up on some stuff and i was reading this blog post about the four new heroes added into captains mode. i have never played either of these heroes in captains mode if they were ever in DotA, but as far as i understand, abaddon has to be the single best trilane support hero in the pool. even with a measly 500 range on his purge, i dont see a single hero that would universally fit so well with practically any single trilane. elder titan should be interesting, too.
i remembered this quote from five weeks ago or so too
do you still think this
could you explain to me why he is not a fantastic support hero in most scenarios, preferably like i am six years old or something
because maybe i am missing something obvious
for your convenience, here are some quotations from the blog post i linked (written by someone way smarter than me):
"support Abbadon is extremely viable", "carry Abaddon doesn’t seem to be working too well", "Semi-carry Abbaddon also appears relatively viable", "there is absolutely no evidence that he’s working as a carry"
to note is that hir list is not normalized for creep kills per minute versus game time, which would cause heroes that want the game to end earlier rather than later to score lower. i guess take it with a bit of salt
Last edited by Longview; 2013-08-23 at 12:28 PM.
So, my last 4 games of dota...
Russian guy picks Pudge ''I mid or feed quit'';
Proceeds to quit anyway, even though he had mid.
Bunch of people that don't speak english at all
A zeus trying to rush basher
and well, that's it....
I played two this morning, my team threw the first one horribly and in the second one the opposing team threw the game really badly so it all evened out.
Hah. I had a game with a Pudge player called "mid or feed"; I told him I didn't give a fuck. He starts flipping the fuck out, apparently not realizing I've no interest of even being mid. After feeding a few times he abandons with his only items being a few Iron Branches. Geez.
I think we actually won that game anyway, but I can't really remember. Well, it was one or the other.
A support Abaddon sits in the back and waits to heal his allies. For this reason, it's pretty easy to simply ignore him completely and just expect to have to do a bit more damage to his mates for kills. He gets zero use out of his ultimate and very little use of frostmourne, just being a weaker version of omniknight. If we're talking strictly trilanes, then obviously Abaddon's two support spells are very useful. I wouldn't call him the best, repel is still a far better spell than Abaddon's entire kit combined and I'm not sure why Omni continues to get flat out ignored by the community when he was a god tier pick in d1.
I would take data like the post you linked with a pile of salt. Trying to find statistical relevance between gold gain and win rate is pretty dangerous in public DotA. This doesn't show you what heroes serve as more stereotypical carry heroes, as the guide suggests, it only shows you what heroes need gold to perform any role at all. This is less because of the hero and the role and more because of the community. Antimage can make a pretty effective tank and initiator if he just stacks health early on (vanguard rush used to be the only way to play him). Now, everyone goes for battlefury no matter how long it takes them to build it. Of course an Antimage with only 300 GPM is going to lose the game when he's at 35 minutes and still only has a perseverance. On the other hand, most Abaddon players will build support items, even if their GPM is really high. A 600 GPM Abaddon is just an Abaddon with mek/pipe/sheep, and of course he's more likely to win the game than an Abaddon with just mek.
Overall, I'd say the chart linked in that blog post is next to meaningless. All it gives us is trends for how people play heroes. The heroes that people INSIST on carrying with are near the top, and the heroes people insist on supporting with are near the bottom, for the most part. Let's observe that Silencer is near the bottom of the list, not too far away from LoA. Does this mean Silencer is a better support than a carry? I don't know a single person in my bracket who thinks Silencer should be a support hero, yet people insist on Abaddon supporting - why?
You don't play Abaddon as a carry because he has super good damage scaling and makes the best use of his gold. You play him as a carry because if the enemies are attacking you instead of your team, you will always know who to shield (yourself). Furthermore, as you soak damage you can simply pop your ultimate mid fight and keep smacking people while you get healed. Abaddon's ability to permanently stick to a target and be impossible to kill is what makes him scary, not his ability to save allies from death. Like I already said, Omniknight does that way better, anyway. The only advantage I give Abaddon over Omni at supporting is purely in trilanes, which is the only circumstance where I would skill death coil on this hero. At the same time, I would still build carry items as I could.
To sum up my opinion, it is that you will see a big change in Abaddon trends, especially with gold use, as people learn how to actually play him. He is such a strong hero that it doesn't matter how you play him in most brackets. His ability to flat out ignore towers with aphotic shield will allow him to win most lanes anyway. However, I invite you to watch this (http://dotabuff.com/matches/283695761) game and see how useless support Abaddon is if the game lasts longer than lane phase. The guide even admits "semi-carry" is viable, which is the role nearly all strength heroes play (or should play). You're not going to rely on LoA to do all the damage, no, so I guess "carry" in that sense is not what I mean. He's your secondary, like Jugg, SB, Wolf, Sven, Siren, Drow (these days), Slardar, Viper, etc.
Last edited by Lysah; 2013-08-23 at 07:11 PM.
A few stats on items for Abaddon this month:
Carry:
Mask of Madness: 358,283
Skull Basher: 291,814
Sange and Yasha: 264,017
Assault Cuirass: 214,921
Heart of Tarrasque: 163,297
Support:
Urn of Shadows: 356,276
Mekansm: 234,645
Arcane Boots: 132,220
I could have included Vladmir's Offering (666,436) as a support item, but the fact is most pub players will buy it on a carry Abaddon anyway.
When only 20,000 more meks are purchased on Abaddon (in a sample of 2 million games) than ACs, I don't think it's fair to say that support is more popular than carry. I know you're going to say that pubs are not a reliable source for this information, but the purpose of this post is mainly to prove that most Abaddon players do not build support items.
By the way, sheepsticks bought on Abaddon this month? 4381. That's a 0.22% build rate.
EDIT: Another interesting stat: Observer wards - winrate on Abaddon = 38%. On Rubick winrate = 34%, on Lion = 37%. Normal winrate = 36%
Well if it's anecdotal evidence that we're going by...
Last edited by mmoc6f376c0517; 2013-08-23 at 08:24 PM.
An enemy team that can't break 1000 gpm combined is a bit TOO anecdotal for me. The game I linked wasn't a complete smash, we actually lost heavily early game as a result of support Abaddon. Luckily, the usefulness of his spells fades very, very quickly.
Even if I believed that people build carry Abaddon more than support (I don't - statistical analysis pulls from all skill brackets and I'm heartlessly not concerned with the 99%) it doesn't change the fact that his win rate is higher with carry items than it is with support items, just browsing dotabuff's stats on him. I don't know what your point is by saying that sheepstick is rarely purchased on him, most support Abaddons never get past mekanism so of course you won't see much else. Half the ones that do will transition into building "carry" items like AC as Necrolyte would.
Well, yeah... carry items always have higher winrates than support items. Take crystal maiden. Winrate with Mekansm = 68%. Winrate with Daedalus = 79%. Abaddon: Winrate with mekansm = 64%. Winrate with Daedalus = 81%. Same pattern for every hero... Lich: Mek = 62% WR, Daedalus = 76%. Naga: Mek = 63% Daedalus = 77%. Doom: Winrate with mek = 53%, Daedalus = 76%. If you're wondering why I chose daedalus, it's because it's pretty much the only item that is good on every single carry.