Last week I crunched the numbers on AM’s itemization, and came to the conclusion that the greedier an Anti-Mage is, the more likely he is to lose. With Battle Fury frequently being a bad choice. This time I am looking at Bounty Hunter’s infamous Battle Fury. The item is widely hated by a large amount of players, and was removed from the recommended items in the New Bloom update. This hugely changed how pub players are itemizing the hero. Note this analysis does not include any 6.81b data.
Win Rate and Items
Bounty Hunter has had a slowly dropping win rate of the past few patches. Dota Version wise, he has been mostly untouched, with just 3 tweaks to Track. The small change in win rate is a tricky factor to find the cause of. I would assume though it’s primarily a result of item changes and hero popularity shifts. The massive decrease in Ember Spirit picks could have had a negative effect on Bounty. As far as items go however, things are very drastic.
As you can see, 6.80 Bounty Hunters had a Battle Fury when the ancient exploded in 38.1% of games. For comparison Desolator had the 43.03% pickup rate, and Battle Fury on Anti-Mage in the same patch had 60.45%. After the New Bloom update this plummeted to only 23%, which is a massive change for pubs. The likely cause is the New Bloom update reworking recommended items. It removed Battle Fury, and now the core Items are Drums and Desolator. Both of which got a boost to their build rate as BF dropped. Additionally Sange and Yasha was removed from the recommended items, and halved in build rate as a result. A few other items were also removed, as can be seen below.
With Urn of Shadows now being listed in the Early Game section instead of Situational, the build rate also went up from about 4% to 12%. While Drums also got moved – to Core – also resulted in a big change. The issue however, is that Bounty Hunter isn’t winning more than before. The likely conclusion is that a massive decrease in Battle Fury on Bounty has had no impact on his win rate. 38% to 23% is a massive decline in build rate, but his win rate actually went down ever so slightly. Why would this be the case? If Battle Fury is as bad as people say it is, then his win rate should have gone up a fair bit since heaps of people are getting it less.
Breaking Things Down
It’s important to break things down in Dota, to understand why something is the way it is. Battle Fury for example gives you 3 things; Sustainability, Damage, and Creep Clearing. The faster Creep Clearing gives an increased farming ability and is why people didn’t like the item on Bounty. People focused solely on the passive cleave that Battle Fury gives, and justified that because the hero is primarily played as a ganking roamer, he has no need for a farming item. This makes sense. But I assume the flaw with this justification is that Bounty Hunters never got Battle Fury to farm, they got it because it was a very convenient all in one package that gave them the regeneration they needed to keep roaming, alongside a respectable 65 damage. Being recommended as core helps as well, but that was the primary reason it was in Core items to begin with. It’s not because of the Jinada rework which occured years and years ago.
The Battle Fury pickup didn’t drastically weaken his early game. A rushed BF or Midas often gives you better farm, at the cost of an better early game. A Lifestealer who rushes Midas has less lane presence and kill potential than one who rushes Phase Boots. But this never seemed to be the case with Bounty; it wasn’t brown boot into BF, it was an eventual BF.
Both in 6.80 and 6.81 it would appear that most Bounties are getting the same thing. They get primarily Damage and Sustainability. Whether the regen is through Battle Fury, Urn, or Vlads, and the Damage through Deso, Mkb, or Battle Fury appears to be mostly irrelevant. They are getting the same thing anyway, they’re just doing it in slightly different ways. There are difference between the items you get, but they are not statistically relevant in the large picture. A large part of the items are still very similar; damage and sustain. There’s no cleave, but there’s Corruption, which does a different thing, but an item is more than the passive/active. They are also likely to be playing the hero the same way, as a ganking roamer. They’re not suddenly getting Mobility items over a Battle Fury, they’re just using alternate items to achieve a very similar outcome.
At the End
Personally I prefer a Bottle, which isn’t suggested for some reason. I think in your average pub Bounty can do more than he can in the pro scene, and thus shouldn’t get a Dagon. I think that a Bottle and some stats, into a damage item and perhaps a BKB is a very solid way to build the hero. Perhaps chuck some utility with a Vlad or Orchid in there somewhere.
From a Statistical stand point though, you would assume if Battle Fury is terrible on Bounty, a massive decrease in the item’s popularity would cause a surge in win rate. But it didn’t. I can thus primarily conclude that although widely seen as not optimal, Battle Fury on Bounty seems to end up doing almost the exact same thing as another item. If you want Damage and Regeneration on Bounty, how you get it is statistically irrelevant to the outcome of your game. I would assume that the more important aspects in playing Bounty, are doing things like moving around the map effectively, and getting frequent high impact Tracks off.
Statistically though, going Battle Fury on Bounty Hunter in a Pub, puts you in no worse of a position than buying the items you would otherwise buy.