Femme Fatale — Medusa and Spectre Trends
Defining what exactly hard carry is in Dota is no easy task—back in the day the playdota forums were filled with discussions who the hardest carries are in the game and as in any online argument no side has ever changed their mind, making definition pretty much irrelevant and highly subjective. For the purpose of this article, we are sticking with a broad understanding of the concept—hard carry is a hero that scales with items better than most other heroes, but starts off weaker than most to balance it out. It was also the case that many hard carries were thought of as one-man-army heroes, which could single-handedly win the game, provided they had enough time to farm.
The development of the game and metagame made the latter concept less relevant, with the average level of play increasing dramatically over the last several years and the 1v5 mindset becoming less and less common. Many changes over this period have balanced heroes out, introduced/altered certain items and overall made it more team-based—AoE/Comeback gold is one of the changes which promotes actual teamwork on a constant basis and can be generally considered healthy, despite an outcry from a very vocal minority. So, with this focus on teamplay, fighting and objectives, is there really a place for heroes, which need to farm for prolonged periods of time (frequently sacrificing early-game) to become effective? As it turns out, at least in this patch the answer is “yes”.
Medusa
We haven’t seen the hero being picked in a serious competitive match for a very long time, and it is unlikely we shall see any development in this regard, especially in a patch where Quas/Wex 
Invoker is so dominant. In public matchmaking, on the other hand, the hero has become one of the best ranged carries — her win rate has increased by almost 4%, putting her at borderline OP 55.36% .
The hero has a lot to offer. She is one of the few heroes who remained relatively unscathed by the Neutral/Ancients changes in terms of flash-farming potential and she also benefitted from a global +30 HP change a lot—her starting HP was among the lowest in the game. Moreover, she has received nothing but buffs 7 patches in a row and ever since the 
Stone Gaze rework the hero has been getting stronger and no global changes have worked directly against her. The last hero with this kind of treatment was 
Lina and we’ve all seen how it turned out—a very small last buff has broken the camel’s back and her popularity exploded. Should we expect the same to happen with Medusa and is 6.86 the last straw?
As an avid fan of the hero, I feel the hero is currently in a very good spot. Her win rate across brackets does not go below 54% until 5k+ games, where it falls down to 52.5%. Her pick rate in the top bracket is rather small, with only a 2.1% chance of her appearing in an actual game. The main reason behind her poor win rate and low pick rate in this bracket is probably the amount of hard counters to the hero, many of which peak in the 5k+ bracket due to a higher skill requirement. Invoker was already mentioned previously, but there are also other counters in 
Nyx Assassin, 
Anti-Mage, 
Phantom Lancer and even 
Lion which make the hero lose half her potency — all of these heroes can decrease her survivability substantially due to mana burn/steal abilities.
What it ultimately makes for is a very strong, but a very situational hero, which should be picked last, if possible. She has hard counters in each role, but if picked at the right time she becomes a better 
Gyrocopter/
Luna in terms of late-game potential. While sacrificing early game capabilities was extremely unwise in the previous patch, currently it has all the chances to work, even if you are facing a very aggressive lineup—just make sure you have enough outpush and turtle yourself to victory.
Spectre
Spectre has a 59.29% win rate on average and almost 55% win rate in the 5k+ skill bracket. This hero is currently the strongest core in the game and an overall second best hero in terms of win rate. How was she under the radar for so long?
The main problem with the hero was her inability to flash-farm. In a game riddled with heroes, which could clear a huge stack for ~600 gold by pressing two buttons, Spectre was heavily out of place — she never had the chance to come online in a proper fashion.
With the current changes and a heavy nerf to the ability of heroes to jungle, she is suddenly much better off. Her laning stage can not be called excellent, but she is unlikely to die, unless she goes for an aggressive build and then makes an aggressive play — her escape is not the most reliable and has a huge mana cost, but she can generally survive any early-game encounters. She can also be quite independent, when facing a solo offlaner for the same reasons. This allows her supports to be more effective and contribute to the success of the team in other places, while she gets direly needed solo XP to reach level 6.
After that her game is completely different — her ability to be in any fight and finish-off fleeing enemies is unparalleled, while her late-gaming scaling is among the highest in the game, if not the highest, since she is one of the few heroes who can utilise 7 or 8 slots without sacrificing much in terms of stats.
Her build was more or less the same for the past several years, with almost no changes in terms of what she builds, but frequent changes in the order. The boots selection is down to personal preference, with 
Phase Boots having a small 1.1% edge over the 
Power Treads. Of the holy trinity of stats items: 
Urn of Shadows, 
Ring of Aquila and 
Drum of Endurance, players generally prefer the second option, but it is not uncommon to see two of these items picked up.
Recent changes to her passive skill—
Dispersion, have also made 
Vanguard one of the better options, especially if you are facing a very aggressive ganking lineup. This item has been booed upon by the community for a very long time, but with a build concentrated on maxing out Dispersion it can be a perfect fit, especially considering how Vanguard’s upgrade 
Crimson Guard is an overall alright item in the late game.
Radiance rush after one stat item is generally the play for a high level Spectre, but there have been games where this item was skipped in favor of an earlier 
Diffusal Blade. While I would generally advise against an early Diffusal, there are games where it might come in extremely handy—strong debuffs/silences do not necessarily destroy Spectre, but they can be a great nuisance on her teammates. Additionally, a reliable instant slow is never a bad thing to have. The DPS from these items cannot be directly compared, however if hitting a single target with an average armor/magic resistance, Diffusal actually wins. It is just a rare occasion for Spectre to be focusing on a single target in the first place.
Conclusion
Overall, with the Invoker nerfs and slow fade of Gyrocopter and 
Windranger, Spectre and Medusa have high chances of becoming the next “cancer” heroes. They are not necessarily annoying to play against, as is the case with previously popular heroes, but if the trends keep their upward slope in terms of pick rate, both of these femme fatales have the potential to spiral out of control, especially if professional players suddenly start picking them.
The big question remains, however—does these trends signal the return of hard carries in general? Perhaps it is a bit too early to try and predict the meta, since it has a tendency to change unpredictably, but as an overall strategy going late-game might finally be worth the price.
There were many changes previously which have addressed reckless push and even though the overall strategy has been buffed in 6.86, it is a far cry from the pushes of the early days.
Flash farming midrange heroes who preferred snowballing to methodical and patient farming are also heavily crippled—there are no more mid-late heroes who can win the game by simply having an economy boost. At the very least, the means of getting these economy boosts very early in the game have been fixed.
This leaves us with the most capricious lot who did not get hit in terms of viability—the hard carries. The heroes who have been largely ignored competitively since TI2 might be on the rise, not only because of getting individual improvements, but because everyone else is worse off. And for anyone missing smart turtle Dota and epic LGD high-ground defenses, this patch might become a rare present.
Ayeee lmao
Also, this comment keeps getting deleted 4hed. Viva le mango.
I would be happy to pick AM every game if these things become the new cancers.
What is happening to mangos?
soon, i am gonna bash people in face for mentioning enchanted mango again
Sorry @Juice - accidentally deleted your comment along with the others.
here is his post for anyone wondering (and he brings up a good point):
Medusa has an enormous bug in 6.86 which I think gonna ruin her shot in pro game play. The Mystic Snake doesn't steal mana from illusions. This is huge for medusa as one of her main counters are illusion based carries. I Guess because Nobody cares about medusa on Reddit it doesn't get any attention/visibility on front page and probably not gonna get fixed soon. it's a shame...
>Invoker
>Nerfs
Done with that "conclusion".
Hate playing with average Medusa/AM on team, 4v5 for 40 min
At least Spectre can help during early/mid
You want to see old faceles void buffed?Yes, with backtrack!
link:https: //mega.nz/#!H9QzkIJT
pasword: !IJWCF0dfFy7ePwTev2rID2OyjVstf5LWwH98S-pMJPw
Pls read and send to icefrog@gmail.com
Hi Skim, thanks for re adding the comment... I just wanted to add - Medusa's win rate on 5k+ mmr is still at the lower side with about 49% in the last month. While spectre's win rate on 5k+ mmr is still decent with about 54%.
This comment was deleted 2 hours ago
Thanks , good info.
@Mogginho if you guys don't support and provide space to the AM, he can't get BF up in like 22mins, or he gets it up before 20min mark but all towers are gone, of course he can't fight. Give enough space and a decent AM will get boots and BF in 15 mins, treads vlads in 18, manta in 25. So the problem always comes from whether you are willing to give AM all these or not. But then again, AM is never really a fighting hero so I'm not sure if you can even compare them like this. AM can start split pushing well with BF vlads, so that's before 20 mins. He just can't be joining actual 10-hero fights.
am can't kill spec i think
15th
The only problem with picking these heroes is that you will always always be dual laned against by an aggressive dual offlane. The exception to that is whenever the enemy pick those heroes in which case your offlane/rs will be ineffective at shutting them down or just feed which explains why you can never pick those heroes and win, yet they still get such high average win rates.
Lol Spectre was always cancer OP hero and every good player knew it, wtf is this article about.
Medusa is my mmr booster hero :P
I continue to encounter Specter every game, I think that this hero will be spammed so much
Medusa has a few advantages over Spectre, tho.
She can absolutely wipe a lane with good Mystic Snakes, which does sick damage early. Most heroes either fall in three categories: those who can nuke her down early and have plenty of mana (Lina or QoP, for instance) lose huge amounts of hp to Snake. Those who can nuke her down and are tanky enough to survive (Sven, Tiny) won't even have the mana to do anything given good Snake harass. And most other heroes lack the nuke potential to get through a mana shield. Alongside a laner with stun+nuke potential, she can get a kill pretty early.
Medusa is also tons better than Spectre in the early-mid game stage. Once she gets Stone Gaze at level 6 (don't level it again until you've fully levelled everything else), she can TP in to defend a tower the enemy is sieging and barring the most perfectly timed stun instantly open with Stone Gaze right out of the teleport, forcing the enemy to back off.
She lanes FAR better than Spectre, teamfights better and farms better. She can come online really early with a quick Maelstrom (altho I religiously build Linken's first, every time), farming jungle stacks with Snake and Split Shot.
^^ While all this is true Medusa has a major disadvantage at the fact that her ult is easy to dodge, Her snake isn't the most effective during teamfights, And her Mana Shield is easily countered by Mana burn or a simple Diffusal blade that is already picked up on a decent amount of heroes.
Spectre on the other hand has an ult like Zeus's but could be used for free map control when she is dominant over the enemy team and in team fights with her ult and Radiance all she has to do is be there and the odds of winning the team fight are improved greatly
@Namer The problem with Medusa is that she is very kiteable. Like it was mentioned, in higher skilled gameplay she has to be last picked or heavily supported by other picks. With huge increase in Invoker picks it is pretty difficult to play effectively, as the notorious EMP and general teamfight control for low mobility heroes. A bkb is an item that you generally don't like to pick up as a Medusa too.
Along with nerfs to Gyro in 6.86, MKB was nerfed to nerf damage but buff on mini-stun proc. This indirectly nerfs Medusa's split shot secondary damage targets.
I've had huge success on Medusa, and I'm an avid fan of maelstrom build.
I believe OD would be a mad matchup too, with the Int steal orb and Ulti draining mana, forcing BKB pick up at some point. Slark is a bad matchup too. A good Slark player will time the dark pact for the Stone gaze really well and continue to pummel you.
Her ulti is insanely good though. The problem with most people is that they don't use it at the right time and not 'baited' to use it too early. You have to use it when they've committed too much as it completely forces the enemy to attempt to disengage (bkb'd and all).
@Namer sure, but the diffusal and mana burn spells are really really strong counters to her. On top of that midgame spectre with radiance cannot be matched by any other carry in terms of game impact.
femmes fatales
as a specter player I'm really happy to see her becoming stronger, as someone who knows how good a specter can be, I'm really scared for her to become popular
Gudddd
@Firefly yes, nerfs. That's right. He got nerfed pretty hard in the latest small patches, but got buffed in 6.86.
Please don’t draw attention to spectre. She will get nerfed again… XD
Also, I dislike rushing radiance against tanky teams. If against tanky and dangerous manfighters, like LS or Wk, the build should be diffusal + manta + Skadi (you may want to get skadi before manta, depends).
The only hero I dislike playing against as Spectre is CK. Radiance is not effective against his tanky illusions, and his illusions counter desolate. So the only thing you can do is kite him until his ulti is down, then quickly clear his manta illusions with help from teammates.
Edit: nice article BTW
choi lai dotka
Currently the Enchanted Mango only has a 31.43% win rate. Together, we as a community, have the ability to fix this problem and end this horrific tragedy. Please help. It is our duty as gamers to give this item the respect it is due . KAPPA
Stone Gaze is actually best used as a counter-initiation. You time it to follow up immediately on their initiation, using Linken's to avoid a stun or silence. Since it's CD is low, you can also use it for escape if you time it well enough and TP out while they're facing away (good players will risk trying to stun you. In that case, cross your fingers).
I usually play in a full party when I'm playing Medusa, so I usually have someone to force staff me into the middle of a teamfight (not that they often do). Medusa also comboes well with heroes like Axe and WW, who can force a direction to face on an enemy hero.
And, yeah, Medusa's weakest phase is her midgame, which is when Spectre is strong. Both of them are strong late-game, and I'd pick Spectre against a lineup that has weaker laning, Medusa against a stronger lane.
Her snake's a laning tool, not a teamfight tool. In teamfights, you just spam it whenever you get the chance.
I haven't matched up with OD in 6.86 yet, I'd probably rotate lane if I did. OD and TA counter Medusa hard in-lane, as does anyone else who can easily dodge the Snake. OD always junked Medusa, tho. Sanity's Eclipse too stronk.
Yeah, Medusa has her counters: mana-burners But Spectre's counter is "Don't let her farm" (easy if you keep hitting their jungle, and Spectre gets easily zoned out in lane, so her Radiance timing is screwed), while Medusa's is mana-burner heroes. What you pick depends on your lineup.
Also even an underfarmed and skinny Medusa has a lot of lategame utility simply by virtue of Stone Gaze.
(I play in 1k and 2k where nobody can counter me unless another lane feeds and AM/PL gets fat. So, take all I say with a little salt).
@GreenToa et al. You mango spammers gotta give it a break. Post your offtopic comments once and then stop. I've been sick of seeing you treating these blog posts like twitch chat.
As Medusa being my main hero I have to say that I do not agree with the statement she has to AFK farm very long, well at least with the build I use (Maelstrom, Yasha, Skadi, Mjolmir, BF, Manta, MKB). She needs some time to farm until she gets her first big item (Skadi, Linkens, stuff like that) But after she does she is a really strong presence in the teamfight, even if only with her ult.
While I do agree she suffers more that Spec in the midgame her farming is a lot easier, while Spec has a lot of issues with it.
While yes Medusa does get countered hard by manaburn, if she can ult in the right moment, it matters not as you have most likely won the teamifight with the ult. And I also hear her snake is weak mid-lategame... I disagree, the snake is not supposed to do dmg mid-lategame but she steals a fifth of a team's mana meaning she fills herself up and deals some dmg. (Altough im in the 2k MMR skill bracket, so, ya know, don't take everything I say for true)
This is a great read, thanks. I think you should have mentioned OD.
While these 2 hard carries seems to be back faceless void has been thrown from a menacing threat to a crappy initiator
Hex on medusa is actually not bad against heroes like antimage. Rapier is overall best item hehe
Nice article
But as for me there are some things you hadn't mentioned.
1) insane lane. Snake is kinda op even on 4k. U can spam it for free and each wave will be cleaned ezly. Moreover it results in a fast push of t1, which is nice for the whole team. Usually my opps just give up on getting farm and stay away from creeps, while i have freefarm by any means.
2) mael is the only build could be used. Getting maxed shield at 9 is essential as with aquilla dusa become extremely tanky and not likely a person midder wanna gank. And on lvl 9 mael compensates her inability to farm jungle. Snake with this mana steal percentage buff makes it really easy to replenish mama in jungle (both satyrs camps, golems, stacks)
3) manta is essential item giving both offensive and defensive stats. Avoiding silence and projectiles and more arrows to shot on every one you see? What can be better?
4) grading mael to m'jolnir is a great boost to her dps, it can be compared to mkb on gyro. More as=more shoots with active ulty, lightnings deal great dmg too on all slowed targets and shield makes her a not desirable aim.
Summing up all of these you may come to an obvious conclusion, that !if not counterpicked! dusa's mid-game is an easy ride and can be considered equal to spectre's.
I dunno about everyone else but I hope this becomes the case. I personally hate high early aggression matches and prefer the more patient and paced games centered around hard carries farming up and then coming out from farming and melting people's faces for knowing when to finally hop in on the action. If this patch becomes a farm hard and win game meta, it will be a godsend.